FREEDOM251

Freedom251 › Text-to-speech tools

Supertonic 3

by Supertonic 3

Free plan AndroidAPIiPhoneLinuxMacSelf-hosted+2

Visit Supertonic 3 Alternatives

Supertonic 3's own site

What Supertonic 3 is

Supertonic 3 is a free, open-weight text-to-speech model from Seoul that runs locally through ONNX Runtime on a CPU. It can synthesize speech in a browser or fully offline on a local machine, without a GPU, cloud connection, or synthesis API. The model supports 31 languages without per-language fine-tuning and accepts expression tags such as <laugh>, <breath>, and <sigh>. It produces 44.1 kHz, 16-bit WAV audio. The package includes fixed preset voice styles; a linked Voice Builder creates custom styles from reference audio, while the archived release itself has no official voice-cloning pipeline. Examples and runtimes are listed for Python, Node.js, web, iOS, Android, and C++. Python components and examples use MIT licensing with attribution, while model weights use OpenRAIL-M restrictions. Commercial use is allowed under those conditions. The repository was archived on September 9, 2026, ending development and support.

Who it suits

It suits developers needing multilingual local or offline speech synthesis on supported runtimes. Commercial users should review OpenRAIL-M restrictions and attribution requirements.

What it does well

  • Runs offline on CPU without cloud services
  • Supports 31 languages
  • Includes expressive speech tags
  • Outputs WAV audio
  • Commercial use permitted with restrictions

What to know first

  • Repository archived; no further support
  • No official voice-cloning pipeline in release
  • OpenRAIL-M has downstream restrictions

Verdict

Supertonic 3 offers local multilingual speech generation with CPU and offline operation. Its archived status and licensing conditions matter for anyone planning ongoing production use.

Facts from Supertonic 3's own pages

API
The maker links a Supertone API for expressive voice generation across 31 languages with adjustable speech controls. · supertonic3.github.io, 12 Oct 2026
API access
Yes · supertonic3.github.io, 20 Sept 2026
API service
The maker describes a Supertone API for expressive voice generation across 31 languages with adjustable speech controls. · supertonic3.github.io, 12 Oct 2026
Browser service
The linked Supertone Play page is described as offering 263 voices across scenes and emotions in a browser without installation. · supertonic3.github.io, 12 Oct 2026
Code license
The Python package, runtime bindings, and example code are MIT-licensed with attribution required. · supertonic3.github.io, 12 Oct 2026
Commercial use
The model weights use OpenRAIL-M, which allows commercial use subject to use restrictions and attribution requirements. · supertonic3.github.io, 12 Oct 2026
Developer runtimes
The maker lists Python, Node.js, web, iOS, Android, and C++ examples and install instructions. · supertonic3.github.io, 12 Oct 2026
Export formats
WAV · supertonic3.github.io, 20 Sept 2026
Free plan
Yes · supertonic3.github.io, 20 Sept 2026
Headquarters
Seoul, Republic of Korea · supertonic3.github.io, 28 Sept 2026
Integrations
The maker lists Python, Node.js, web, iOS, Android, and C++ runtimes with examples. · supertonic3.github.io, 12 Oct 2026
Languages
The model supports 31 languages without per-language fine-tuning. · supertonic3.github.io, 12 Oct 2026
Licensing limit
The maker states that OpenRAIL-M has downstream use restrictions and is not equivalent to MIT. · supertonic3.github.io, 12 Oct 2026
Local use
The maker says the model runs in a browser or fully offline on a machine, without a GPU, cloud service, or API. · supertonic3.github.io, 12 Oct 2026
Model license
The model weights use OpenRAIL-M, which the maker says permits commercial use subject to use-based restrictions and attribution. · supertonic3.github.io, 12 Oct 2026
Model limitation
The archived repository says its open-weight release focuses on fixed-voice local TTS and does not include an official voice-cloning pipeline. · github.com, 12 Oct 2026
Offline use
The model can run in a browser or fully offline on a local machine without a GPU, cloud connection, or API for synthesis. · supertonic3.github.io, 12 Oct 2026
Output
The archived repository says the model outputs 44.1 kHz, 16-bit WAV audio. · github.com, 12 Oct 2026
Paid service mention
The maker says a Pro Play Desktop subscription unlocks commercially licensed voices, zero-shot cloning, and unlimited usage. · supertonic3.github.io, 12 Oct 2026
Performance claim
In the maker's benchmark, Supertonic 3 produced audio at 5.00 times real time on a 16-thread CPU across 30 samples. · supertonic3.github.io, 12 Oct 2026
Platforms
web,ios,android,api · supertonic3.github.io, 20 Sept 2026
Preset voices
Supertone Play offers 263 voices across scenes and emotions and lets users try Supertonic 3 in a browser without installation. · supertonic3.github.io, 12 Oct 2026
Product
Supertonic 3 is a 99M-parameter open-weight multilingual text-to-speech model that runs locally on CPU using ONNX Runtime. · supertonic3.github.io, 12 Oct 2026
Purpose
Supertonic 3 is a 99M-parameter open-weight text-to-speech model that runs locally on CPU through ONNX Runtime. · supertonic3.github.io, 12 Oct 2026

Where Supertonic 3 stands on Freedom251

Look at these too